Building reputation on the syndicated lending market
-
Upload
christophe-j-godlewski -
Category
Economy & Finance
-
view
84 -
download
0
description
Transcript of Building reputation on the syndicated lending market
Building reputation on the syndicated lending marketA participant bank perspective
Marie-Hélène Broihanne
UniStra, EM Strasbourg (LaRGE Research Center)
Christophe J. Godlewski
UHA & EM Strasbourg (LaRGE Research Center)
AFFI Conference 2014, Aix-en-Provence1
Take away
• Role of lead bank’s reputation for the syndicate participants
• Measures of reputation using Bloomberg league tables
• Focus on reputation capital, relationship capital, expertise capital…
• + 4 600 loans in 28 European countries• First deal with a « top-dog » is crucial for reputation• No accumulation effect• Relationship & expertise matter but much less
2
Background
• Major banking scandals• Bank reputation crucial for activity, performance…• Focus on syndicated loan market• Banking pool = lead bank + participants• Leaders = diligence, allocation, monitoring…• Participants rely on lead reputation• Global syndicated market relies on participants for funding• Focus on Europe: private credit; underwritten deals; regional
credit markets; « small world »• Theory: repeated interactions + collaboration + knowledge
transfers => reputation => organization of pools + design of contracts (Pichler & Wilhelm 2001, Tykvova 2007)
• Scarce empirical evidence (Cai 2010, Cai et al. 2010)
3B a c k g r o u n d | E m p i r i c a l d e s i g n | R e s u l t s | D i s c u s s i o n
Empirical design | Sample
• Bloomberg Professional Terminal Server• Banking pool + loan characteristics at origination• 4 629 loans; 3 506 participants; 28 European countries• DE + ES + FR + NL + UK = 65% of sample• Time span: 1/1/1999-31/12/2009• Lead bank = Loan Agent (title/role)• Use BBG rankings 1999-2009• Identify 623 leaders• Focus on top 20 (63% & 79% of total issues & amount resp.)• Hyp.: leader rank (reputation) => syndicate / participant
reputation• Ordered var.: Participant reputation (1-4) = aggregation of BBG
ranks 20-1 (rank < 20 => PR = 0)
4B a c k g r o u n d | E m p i r i c a l d e s i g n | R e s u l t s | D i s c u s s i o n
Empirical design | Variables
5B a c k g r o u n d | E m p i r i c a l d e s i g n | R e s u l t s | D i s c u s s i o n
Proxy Variable Mean Std dev. MedianParticipant reputation 1.28 1.53 0.00
Prestige
First deal with top 3 leader 0.01 0.10 0.00
First deal with top 20 leader 0.03 0.18 0.00Deals with top 3 leader 8.94 14.50 3.00Deals with top 20 leader 24.06 33.10 8.00
Relationship Same leader 0.16 0.37 0.00
ExpertiseSame loan purpose 0.56 0.50 1.00Same industry sector 0.64 0.48 1.00Same country 0.52 0.50 1.00
Activity
Q1 loan amount 0.20 0.40 0.00Q5 loan amount 0.20 0.40 0.00Q1 syndicate 0.20 0.40 0.00Q5 syndicate 0.19 0.39 0.00
Controls
Syndicate 25.86 36.54 17.00Loan amount 1761.00 6813.00 600.00Spread 122.44 116.35 75.00Maturity 5.72 3.47 5.23Collateral 0.20 0.40 0.00Covenants 0.20 0.40 0.00
Results
6B a c k g r o u n d | E m p i r i c a l d e s i g n | R e s u l t s | D i s c u s s i o n
Ordered probit regressions. Y = Participant reputation. Controls: syndicate size; loan amount, spread, maturity, purpose, currency; collateral; covenants; term; tranches; year; country; SIC.
First deal with top 20 leader 0.8252***
(0.0331)
First deal with top 10 leader 1.0599***
(0.0390)
First deal with top 5 leader 1.4260***
(0.0516)
First deal with top 3 leader 7.2629***
(0.0799)
N. clusters 3151 3151 3151 3151
log likelihood -48427.6476 -48359.8737 -48314.8890 -48169.9346
Chi² 82755.2081 81096.6357 83680.9866 106007.6572
Pseudo. R² 0.0688 0.0701 0.0709 0.0737
Results (cont.)
7B a c k g r o u n d | E m p i r i c a l d e s i g n | R e s u l t s | D i s c u s s i o n
First deal with top 20 leader
0.8889*** 0.8967*** 0.8973*** 0.8954***
First deal with top 10 leader
1.1269*** 1.1342*** 1.1349*** 1.1337***
Deals with top 20 leader
0.0039*** 0.0039***
Deals with top 10 leader
0.0050*** 0.0049***
Deals with top 5 leader 0.0079*** 0.0078***
Deals with top 3 leader 0.0093*** 0.0093***
First deal with top 5 leader
1.4870*** 1.4925*** 1.4934*** 1.4920***
First deal with top 3 leader
7.3211*** 7.3279*** 7.3279*** 7.3258***
Deals with top 20 leader
0.0037*** 0.0037***
Deals with top 10 leader
0.0048*** 0.0047***
Deals with top 5 leader 0.0076*** 0.0075***
Deals with top 3 leader 0.0091*** 0.0089***
Results (cont.)
8B a c k g r o u n d | E m p i r i c a l d e s i g n | R e s u l t s | D i s c u s s i o n
First deal with top 20 leader
0.9446***
Deals with top 20 leader 0.0040***
First deal with top 3 leader
7.3570***
Deals with top 3 leader 0.0092***
Same leader 0.2628*** 0.2428***
Results (cont.)
9B a c k g r o u n d | E m p i r i c a l d e s i g n | R e s u l t s | D i s c u s s i o n
First deal with top 20 leader
0.9253*** 0.9343*** 0.9168*** 0.9644***
Deals with top 20 leader
0.0037*** 0.0036*** 0.0038*** 0.0034***
First deal with top 3 leader
7.3456*** 7.3483*** 7.3293*** 7.3662***
Deals with top 3 leader 0.0087*** 0.0086*** 0.0088*** 0.0084***
Same loan purpose 0.0738*** 0.0569*** 0.0443*** 0.0339**
Same industry sector 0.0898*** 0.0725*** 0.0539*** 0.0437***
Same country 0.0565*** 0.0384*** 0.0263** 0.0139
Results (cont.)
10B a c k g r o u n d | E m p i r i c a l d e s i g n | R e s u l t s | D i s c u s s i o n
First deal with top 20 leader
0.8379*** 0.8367*** 0.8305***
Deals with top 20 leader
0.0249*** 0.0245*** 0.0246***
First deal with top 3 leader
7.3021*** 7.3234*** 7.3075***
Deals with top 3 leader
0.0090*** 0.0088*** 0.0089***
Q1 loanamount
-0.1545*** -0.1835*** -0.1686*** -0.2058***
Q1 syndicate 0.1370*** 0.0930** 0.1346*** 0.0948***
Q5 loanamount
0.2558*** 0.2477*** 0.2483*** 0.2348***
Q5 syndicate 0.0227 0.0442** 0.0711*** 0.0971***
Results (cont.)
11B a c k g r o u n d | E m p i r i c a l d e s i g n | R e s u l t s | D i s c u s s i o n
• Full specification• All coefficients are significant & robust with similar m.e.• Robustness checks• Different sub-samples with more “problematic” deals• Short maturity; no collateral; no covenants…• Most of results remain robust• Different levels of participant activity; time frame;
geographic area• Most of results remain robust
Discussion
• Reputation of financial institutions• Focus on participant banks on the European
syndicated lending market• Prestige of the arranger in first deal matters most
(+86%)• Less prestigious leader => +30%• Number of deals has a residual impact• Relationship; large syndication; expertise
(purpose & industry) => +5% to +10%
12B a c k g r o u n d | E m p i r i c a l d e s i g n | R e s u l t s | D i s c u s s i o n